GridapDistributed.jl
GridapDistributed.jl copied to clipboard
Parallel distributed-memory version of Gridap
@santiagobadia I have added compatibility with 32-bit machines. This compatibility inherits from Gridap. It allows keeping 32-bit tests in repositories depending on GridapDistributed, e.g., GridapEmbeded.
Sister PR for https://github.com/gridap/Gridap.jl/pull/978
Here are some notes on the issue: - In some specific scenarios (which are not as rare), owned rows of the matrix contain column identifiers from dofs which are NOT...
I implemented support for Dirac deltas, similar to the one in Gridap.jl but restricted to the definition by tags. The definition at arbitrary points will require a parallel search as...
- [ ] Add documentation on assembly strategies - [ ] Add a new assembly strategy that works like FullyAssembledRows but doe not optimize column PRanges.
I'm trying to parallelize a linear elastic problem solver with a concentrated load applied on the middle point of ine boundary. In series code the load is described using DiracDelta...
The error is `ERROR: Scalar indexing on DebugArray is not allowed for performance reasons.` and it occurs at [this line](https://github.com/gridap/GridapDistributed.jl/blob/fac92506e1f5945c4ea12bf41f8ba6f9da382aa8/test/PLaplacianTests.jl#L13) With Julia 1.9.3 it [runs sucessfully](https://github.com/gridap/GridapDistributed.jl/actions/runs/5947075300) The reason for this...
In PR https://github.com/gridap/GridapDistributed.jl/pull/115, we introduced in GridapDistributed a variant of PArrays `assemble_coo!` named `assemble_coo_with_column_owner!` which also exchanges the process column owners of the entries. I guess that it would be...
This allows to get updates for GitHub actions automatically. I have used this for my own packages, the [Trixi.jl framework](https://github.com/trixi-framework), and the [SciML organization](https://github.com/SciML). After merging this, you could also...
I'm trying to evaluate my distributed solution at given x,y coordinates. In series the code works, but in parallel I cannot. I think the error is caused by the fact...